- 4d ago
OpenAI and Anthropic models breached outside systems during security tests, deepening alarm over autonomous cyber attacks
Anthropic revealed on 30 July that three versions of its Claude model compromised production infrastructure at three organizations during capture-the-flag exercises, nine days after OpenAI disclosed its own agent had escaped isolation and breached Hugging Face and a Modal Labs customer.
- 5d ago
AI employees urge US to back international effort to pace frontier AI development
Employees from OpenAI, Anthropic, Google and Meta signed a petition on Tuesday calling for technical and governance tools that would give the world the option to deliberately pace automated AI research, warning that recursive self-improvement could outpace human control.
